Starting from known p38α mitogen-activated protein kinase (MAPK) inhibitors, a series of inhibitors of the c-Jun N-terminal kinase (JNK) 3 was obtained. Altering the substitution pattern of the pyridinylimidazole scaffold proved to be effective in shifting the inhibitory activity from the original target p38α MAPK to the closely related JNK3. In particular, a significant improvement for JNK3 selectivity could be achieved by addressing the hydrophobic region I with a small methyl group. Furthermore, additional structural modifications permitted to explore structure-activity relationships. The most potent inhibitor 4-(4-methyl-2-(methylthio)-1H-imidazol-5-yl)-N-(4-morpholinophenyl)pyridin-2-amine showed an IC50 value for the JNK3 in the low triple digit nanomolar range and its binding mode was confirmed by X-ray crystallography.
Starting from known p38α mitogen-activated protein kinase (MAPK) inhibitors, a series of inhibitors of the c-Jun N-terminal kinase (JNK) 3 was obtained. Altering the substitution pattern of the pyridinylimidazole scaffold proved to be effective in shifting the inhibitory activity from the original target p38α MAPK to the closely related JNK3. In particular, a significant improvement for JNK3selectivity could be achieved by addressing the hydrophobic region I with a small methyl group. Furthermore, additional structural modifications permitted to explore structure-activity relationships. The most potent inhibitor 4-(4-methyl-2-(methylthio)-1H-imidazol-5-yl)-N-(4-morpholinophenyl)pyridin-2-amine showed an IC50 value for the JNK3 in the low triple digit nanomolar range and its binding mode was confirmed by X-ray crystallography.
The mitogen-activated
protein kinases (MAPKs) represent a family
of enzymes involved in several signal transduction pathways, whose
activation is part of a phosphorylation cascade triggered by diverse
extracellular stimuli. Among the members of this family, the c-Jun
N-terminal kinases (JNKs) mostly respond to a variety of stress stimuli
such as radiation, osmotic or heat shock, oxidative insult, and proinflammatory
cytokines, modulating responses such as cell survival and apoptosis.[1] The JNK subfamily is encoded by the three genes jnk1, jnk2, and jnk3,
which in turn give rise to 10 different isoforms through alternative
splicing.[2] Despite their structural homology
and the partially functional redundancy, these isoforms follow a different
tissue distribution pattern, JNK3 being restricted to the central
nervous system, heart, and testis oppositely to the ubiquitous expression
of JNK1 and 2.[2,3] In addition to this, a different
substrate specificity of the JNK1, 2, and 3 suggests the existence
of isoform-specific roles of these enzymes, which were partially disclosed
through gene knock-out studies.[4] There
is well-documented evidence for the critical role of the JNK subfamily
members in several neurodegenerative diseases such as Parkinson’s
and Alzheimer’s disease, as well as in neuronal death derived
by stroke and ischemia/reperfusion injury.[3−6] Furthermore, some members of the
JNKs are also involved in metabolic and inflammatory diseases, and
several studies suggest that these kinases might contribute to the
development and diffusion of some forms of cancer,[7−9] thus emerging
as particularly attractive drug targets. Despite the intense endeavor
in the research of JNK inhibitors, only a scarce number of candidates
have reached clinical trial phases and to date, none of them have
been approved.[10−12] Until early 2010s, a major challenge in the development
of JNK inhibitors has been the achievement of selectivity over the
closely related p38α MAPK,[11] a member
of the same family which, analogously to the JNKs, participates in
regulating the cellular response to stress stimuli. This protein kinase
was also shown to assume a key function in different inflammatory
and neurodegenerative diseases[13−15] and the simultaneous inhibition
of JNK and p38α MAPK is assumed to obtain a synergistic effect
in the treatment of some pathological conditions.[16] Nevertheless, obtaining a JNK-selective inhibitor would
be beneficial to fully elucidate the effective role of this protein
kinase in the aforementioned pathological conditions and thereby assess
its therapeutic potential. Furthermore, most of the reported clinical
trials on selective p38α MAPK inhibitors have been discontinued
because of the insurgence of adverse effects mostly related to liver
toxicity,[17] leading to the assumption the
activity on the p38α MAPK to be undesired for an improved safety
profile of JNK inhibitors.Regarding the selectivity within
the JNK subfamily, the achievement
of JNK isoform-selective inhibitors would be desirable to dissect
the contribution of the different isoforms in various pathological
conditions. However, the JNK1, 2, and 3 share more than 80% sequence
identity, making the development of isoform-specific inhibitors extremely
challenging.In the last decades, pyridinylimidazoles have encountered
a remarkable
success in the field of p38α MAPK inhibition. This class of
inhibitors counts a large number of examples starting from the precursor SB203580 to the optimized compound LN950 (Figure ), until reaching
derivatives with low single digit nanomolar IC50 values
(a review on this class of compounds has recently been published).[18] As can be seen from Figure , the reported p38α MAPK inhibitors
are also able to inhibit the JNK3 with IC50 values in the
submicromolar range, thus offering a suitable starting point for optimization
when aiming to target this enzyme. In 2016, we published compound 1a as a balanced dual JNK3/p38α MAPK inhibitor, which
served as a precursor for the synthesis of a fluorescent probe used
in fluorescence polarization-based binding assays.[19,20] As it is evident from the biological activity of 1a in comparison to the activity of previous inhibitors, modifying
the substitution pattern around the pyridinylimidazole scaffold can
contribute to a shift in selectivity toward the JNK3.
Figure 1
Tri- and tetrasubstituted
pyridinylimidazoles. Data are taken from
Ansideri et al.[19] and Muth et al.[21]
Tri- and tetrasubstituted
pyridinylimidazoles. Data are taken from
Ansideri et al.[19] and Muth et al.[21]Some of us have recently reported the optimization of compound 1a following a covalent inhibition approach (compound 1b), which was based on the introduction of an electrophilic
moiety able to target a noncatalytic cysteine of the JNK3 that is
not conserved in the closely related p38α MAPK.[21] The aim of the herein presented work consists instead in
the achievement of a potent and selective JNK inhibitor by structural
modification of the pyridinylimidazole scaffold following the canonical
concept of reversible inhibition.
Results and Discussion
Chemistry
Despite the overall similarity of their structures,
the herein reported compounds were synthesized following considerably
diverse routes, especially with regard to the construction of the
five-membered heterocyclic central core. The synthesis of compounds 5 and 8 was achieved as displayed in Scheme . The route leading
to the common intermediate 3, starting from 2-fluoro-4-methylpyridine
(2), is based on the Marckwald imidazole synthesis[22] and was previously reported by Laufer et al.[23] The substitution on the imidazole-C2-S position
was obtained by reacting imidazole-2-thione 3 with the
appropriate alkyl halide. Finally, the introduction of the 4-morpholinoaniline
moiety was carried out through nucleophilic aromatic substitution
in acidic conditions, this representing the final step for most of
the herein presented compounds. Applying these conditions to the hydroxyethyl
derivative 6 unexpectedly yielded imidazol-2-one 8, instead of imidazole 7, as a result of a previously
described rearrangement.[24]
Scheme 1
Synthesis
of Imidazole 5 and Imidazol-2-one 8
Reagents and conditions: (a)
four-step route reported by Laufer and co-workers;[23] (b) MeI, K2CO3, MeOH, rt, 18 h; (c)
2-bromoethyl acetate, t-BuONa, MeOH, 55 °C,
3 h; and (d) 4-morpholinoaniline, 1.25 M HCl in EtOH, n-BuOH, 180 °C, 16 h.
Synthesis
of Imidazole 5 and Imidazol-2-one 8
Reagents and conditions: (a)
four-step route reported by Laufer and co-workers;[23] (b) MeI, K2CO3, MeOH, rt, 18 h; (c)
2-bromoethyl acetate, t-BuONa, MeOH, 55 °C,
3 h; and (d) 4-morpholinoaniline, 1.25 M HCl in EtOH, n-BuOH, 180 °C, 16 h.The preparation
of 2,4,5-trisubstituted imidazole 13 and of 4,5-disubstituted
imidazoles 14 and 18a–l is outlined
in Scheme . The route
providing α-diketone 10 starting from
2-fluoro-4-methylpyridine (2) was recently described
by Ansideri et al.,[19] whereas the synthesis
of intermediates 16a–l was achieved following
a similar approach. Ethanones 15a–l were obtained
by condensation of the appropriate ethyl ester with 2-chloro-4-methylpyridine
(9) and were subsequently oxidized by SeO2 to the corresponding diketones (16a–l). Microwave-assisted
cyclization with formaldehyde and NH4OAc in Radzisewski
conditions[25] then afforded the disubstituted
imidazoles 12 and 17a–l, whereas
propionaldehyde and methanolic NH3 were employed to obtain
the 2-ethylimidazole 11. Finally, introduction of the
4-morpholinoaniline moiety at the pyridine-C2 position, giving the
final compounds 13, 14, and 18a–l, was accomplished by the aforementioned nucleophilic aromatic substitution.
Scheme 2
Synthesis of 4,5-Disubstituted Imidazoles 13, 14, and 18a–l
Reagents
and conditions: (a)
route reported by Ansideri et al.;[19] (b)
HCHO(aq), NH4OAc, AcOH, 180 °C microwave
irradiation, 2–5 min; (c) propionaldehyde, 7 M NH3 in MeOH, 80 °C, 4 h; (d) 4-morpholinoaniline, 1.25 M HCl in
EtOH, n-BuOH, 180 °C 16 h; (e) ethyl arylcarboxylate
or ethyl alkylcarboxylate, NaHMDS, dry THF, 0 °C 1–5 h;
and (f) SeO2, AcOH, 70 °C, 2–3 h; (R2 = see Table ).
Synthesis of 4,5-Disubstituted Imidazoles 13, 14, and 18a–l
Reagents
and conditions: (a)
route reported by Ansideri et al.;[19] (b)
HCHO(aq), NH4OAc, AcOH, 180 °C microwave
irradiation, 2–5 min; (c) propionaldehyde, 7 M NH3 in MeOH, 80 °C, 4 h; (d) 4-morpholinoaniline, 1.25 M HCl in
EtOH, n-BuOH, 180 °C 16 h; (e) ethyl arylcarboxylate
or ethyl alkylcarboxylate, NaHMDS, dry THF, 0 °C 1–5 h;
and (f) SeO2, AcOH, 70 °C, 2–3 h; (R2 = see Table ).
Table 2
Effect of Different Aryl and Alkyl
Substituents at the Imidazole C4(5) Position
IC50 values are the mean
of three experiments.
Percent
inhibition at indicated
concentration.
The synthesis of 4,5-disubstituted pyridinylimidazoles 38 and 39, featuring a linear alkyl group at
the imidazole-C4
position, required a different strategy than the examples having aromatic
or branched aliphatic moieties (14 and 18a–l). This was mainly due to the fact that alkyl esters of linear alkanoic
acids did not undergo condensation with 2-chloro-4-methylpyridine
(9) to give the desired ethanone intermediates.An alternative approach to compounds 38 and 39 could also be employed for the synthesis of the 2,4(2,5)-disubstituted
imidazole 43 as well as for the 2,4,5-trisubstituted
imidazoles 44 and 45 (Scheme ). This route started from the commercially
available 1-(2-chloropyridin-4-yl)ethan-1-one (21) or
from the acylpyridines 22 and 23, which
were synthesized by Grignard reaction of the appropriate alkylmagnesium
bromide with Weinreb amide 20.
Scheme 3
Synthesis of Imidazoles 38, 39, 43–45, and 47
Reagents and conditions: (a)
SOCl2, reflux temperature, 5 h; (b) N,O-dimethylhydroxylamine hydrochloride, Et3N,
dry DCM, 16 h; (c) EtMgBr or n-PrMgBr, dry THF, −10
°C, 1–3 h; (d) NH2OH·HCl, 20% NaOH(aq), MeOH, H2O, 0 °C, 1–2 h; (e) TsCl,
pyridine, rt, 24–72 h; (f) EtOHabs, K, 0 °C,
2–16 h; (g) concd HCl, 50 °C, 1–4 h; (h) KSCN,
MeOH, reflux temperature, 4 h; (i) H2O2, AcOH,
rt, 15 min; (j) MeI, t-BuONa, MeOH, 50 °C, 0.5–3
h; (k) 4-morpholinoaniline, 1.25 M HCl in EtOH, n-BuOH, 180 °C, 16 h; and (l) cyanamide, EtOH, reflux temperature,
2 h.
Synthesis of Imidazoles 38, 39, 43–45, and 47
Reagents and conditions: (a)
SOCl2, reflux temperature, 5 h; (b) N,O-dimethylhydroxylamine hydrochloride, Et3N,
dry DCM, 16 h; (c) EtMgBr or n-PrMgBr, dry THF, −10
°C, 1–3 h; (d) NH2OH·HCl, 20% NaOH(aq), MeOH, H2O, 0 °C, 1–2 h; (e) TsCl,
pyridine, rt, 24–72 h; (f) EtOHabs, K, 0 °C,
2–16 h; (g) concd HCl, 50 °C, 1–4 h; (h) KSCN,
MeOH, reflux temperature, 4 h; (i) H2O2, AcOH,
rt, 15 min; (j) MeI, t-BuONa, MeOH, 50 °C, 0.5–3
h; (k) 4-morpholinoaniline, 1.25 M HCl in EtOH, n-BuOH, 180 °C, 16 h; and (l) cyanamide, EtOH, reflux temperature,
2 h.Formation of the corresponding oximes 24–26 and following tosylation of the hydroxyl groups
led to intermediates 27–29. Tosylated oximes 27–29 were
then first converted into the α-aminoketones 30–32 through Neber rearrangement[26] and subsequently
cyclized by KSCN, yielding imidazole-2-thione derivatives 33–35. From these intermediates, it was possible to achieve the disubstituted
imidazoles 36 and 37 by oxidative desulfurization[27] as well as the 2-methylsulfanylimidazoles 40–42 via monomethylation. Alternatively, compound 46 displaying a 2-aminoimidazole core could be prepared by
cyclization of the α-aminoketone 31 with cyanamide.
Intermediates 36, 37, 40–42, and 47 were then reacted with 4-morpholinoaniline,
as previously mentioned, to afford the final compounds 38, 39, 43–45, and 47, respectively.Several analogues of compound 44 featuring a different
substituent at the pyridine-C2 position (compounds 48a–h and 48m, Scheme ) could be prepared by nucleophilic aromatic substitution
of synthone 41 with p-phenylendiamine,
1-phenylethanamine, or with diverse branched or cycloalkyl amines.
In addition, compound 48h and the previously reported 48m(21) were coupled with different
acid chlorides or anhydrides to obtain the corresponding amides 48i–l and 48n–q (Scheme ).
Scheme 4
Synthesis of 4(5)-Methyl-2-methylsulfanyl-5-(4)pyridin-4-ylimidazoles 48a–q
Reagents and conditions: (a)
cycloalkylamine (NEAT or n-BuOH), 180 °C, 24–72
h; (b) p-phenylendiamine, 1.25 M HCl in EtOH, n-BuOH, 180 °C, 16 h; (c) trans-diaminocyclohexane, n-BuOH, 180 °C, 72 h; and (d) acyl chloride or anhydride,
dry pyridine, rt, 16 h; (R1, R2 = see Table ).
Synthesis of 4(5)-Methyl-2-methylsulfanyl-5-(4)pyridin-4-ylimidazoles 48a–q
Reagents and conditions: (a)
cycloalkylamine (NEAT or n-BuOH), 180 °C, 24–72
h; (b) p-phenylendiamine, 1.25 M HCl in EtOH, n-BuOH, 180 °C, 16 h; (c) trans-diaminocyclohexane, n-BuOH, 180 °C, 72 h; and (d) acyl chloride or anhydride,
dry pyridine, rt, 16 h; (R1, R2 = see Table ).
Table 6
Influence of Substituents at the Pyridine-C2
Position
IC50 values are the mean
of three experiments.
Percent
inhibition at indicated
concentration.
According
to the ZINC patterns tool,
compound 48m represents a potential pan-assay interference
compound. However, this compound was synthesized as the intermediate
for the preparation of inhibitors 48n–q. To estimate
the impact of the amide moiety present in compounds 48n–q on the inhibition of the two kinases, the activities of 48m are listed in this table.
The introduction of a methyl substituent on the imidazole-N atom,
providing 1,2,4,5-tetrasubstituted imidazoles 50 and 57, required a distinct approach depending on the desired
N-methylated regioisomer. In fact, double nucleophilic substitution
of imidazole-2-thione 34 using excess of methyl iodide
almost exclusively afforded the regioisomer bearing the substituent
on the N atom away from the pyridine ring (49, Scheme ). The regioselectivity
of the methylation reaction was confirmed by crystal structure analysis
of intermediate 49 (see Figure S1 in the Supporting Information) and was attributed to
the lower steric hindrance offered by the methyl group compared to
the pyridine ring. The regioisomer 54, having the methyl
group on the N atom adjacent to the pyridine ring, was instead achieved
by cyclizing the α-aminoketone 31 with methyl isothiocyanate,
followed by methylation of the sulfur of the resulting N1-methylimidazole-2-thione 51.
Scheme 5
Synthesis of Tetrasubstituted
Imidazoles 50 and 57–59
Reagents and conditions: (a)
KSCN, MeOH, reflux temperature, 4 h; (b) MeI, t-BuONa,
MeOH, 80 °C, 3 h; (c) 4-morpholinoaniline, Pd2(dba)3, Xantphos, Cs2CO3, dry 1,4-dioxane,
100 °C, 18 h; (d) alkyl isothiocyanate, Et3N, 60 °C,
16 h; (e) AcOH, 80 °C, 1 h; (f) MeI, t-BuONa,
MeOH, 50 °C, 30 min; and (g) 4-morpholinoaniline, Pd2(dba)3, XPhos, Cs2CO3, dry 1,4-dioxane,
100 °C, 16 h.
Synthesis of Tetrasubstituted
Imidazoles 50 and 57–59
Reagents and conditions: (a)
KSCN, MeOH, reflux temperature, 4 h; (b) MeI, t-BuONa,
MeOH, 80 °C, 3 h; (c) 4-morpholinoaniline, Pd2(dba)3, Xantphos, Cs2CO3, dry 1,4-dioxane,
100 °C, 18 h; (d) alkyl isothiocyanate, Et3N, 60 °C,
16 h; (e) AcOH, 80 °C, 1 h; (f) MeI, t-BuONa,
MeOH, 50 °C, 30 min; and (g) 4-morpholinoaniline, Pd2(dba)3, XPhos, Cs2CO3, dry 1,4-dioxane,
100 °C, 16 h.This approach, adapting
a procedure published by Xi et al.,[27] represents
an unusual route to tetrasubstituted
pyridinylimidazoles and was recently reported by some of us for the
preparation of tetrasubstituted imidazoles bearing two aromatic moieties
at the 4 and 5 positions.[28] The same method
could also be employed, using the appropriate alkyl isothiocyanate,
to achieve the N-ethyl- and the N-cyclopropyl-imidazole derivatives 55 and 56, respectively. Unlike the majority
of the reported compounds, the introduction of the 4-morpholinoaniline
moiety, yielding compounds 50 and 57–59, was carried out by palladium-catalyzed Buchwald–Hartwig
aryl amination.The synthesis of the 1,5-disubstituted imidazole 66, bearing an aromatic substituent on the imidazole-N1 atom,
was performed
starting from 2-bromoisonicotinaldehyde (60) via a two-step
procedure as depicted in Scheme . Such a route entails the formation of the imine derivative 61 and its direct cyclization through the Van Leusen reaction[29] using toluene sulfonylmethylisocyanide (TOSMIC)
and K2CO3. The analogous route was unfortunately
not accessible for the synthesis of the N1-methyl substituted derivative 67 because of the instability of the corresponding imine intermediate.
As an alternative, the preformed N1-methyl imidazole group was introduced
through Suzuki cross-coupling reaction[30] between 5-bromo-1-methyl-1H-imidazole (62) and pyridinyl-boronic acid 63 (Scheme ). The last step of both routes consisted
of the introduction of the 4-morpholinoaniline moiety. In the case
of the 4-fluorophenyl derivative 64, this was performed
by Buchwald–Hartwig amination giving compound 66, whereas the acid-catalyzed nucleophilic aromatic substitution was
employed for the synthesis of compound 67.
Scheme 6
Synthesis
of 1,5-Disubstituted Imidazoles 66 and 67
Reagents and conditions: (a)
4-fluoroaniline, AcOH, EtOH, reflux temperature, 2 h; (b) TOSMIC,
K2CO3, MeOH/dimethoxyethane 2:1, reflux temperature,
3 h; (c) Pd(PPh3)4, Cs2CO3, H2O, DMF, 60 °C, 24 h; and (d) 4-morpholinoaniline, t-BuONa, Pd2(dba)3, BINAP, toluene,
80 °C, 3 h; (e) 4-morpholinoaniline, 1.25 M HCl in EtOH, n-BuOH, 180 °C, 16 h.
Synthesis
of 1,5-Disubstituted Imidazoles 66 and 67
Reagents and conditions: (a)
4-fluoroaniline, AcOH, EtOH, reflux temperature, 2 h; (b) TOSMIC,
K2CO3, MeOH/dimethoxyethane 2:1, reflux temperature,
3 h; (c) Pd(PPh3)4, Cs2CO3, H2O, DMF, 60 °C, 24 h; and (d) 4-morpholinoaniline, t-BuONa, Pd2(dba)3, BINAP, toluene,
80 °C, 3 h; (e) 4-morpholinoaniline, 1.25 M HCl in EtOH, n-BuOH, 180 °C, 16 h.The 1,2-disubstitutedimidazole derivative 71 was
obtained starting from 2-chloroisonicotinonitrile (68), which was initially reacted in a one-pot procedure described by
Voss et al.[31] (Scheme ). This reaction involves the formation of
an imidate, followed by substitution with acetal-protected aminoacetaldehyde
and final ring closure by deprotection, affording 2-(pyridine-4-yl)imidazole 69 in good yield. At last, N-methylimidazole 70 was obtained by nucleophilic substitution with methyl iodide
and subsequently reacted with 4-morpholinoaniline as previously discussed,
yielding compound 71.
Scheme 7
Synthesis of Imidazol-2-yl Pyridine
Derivative 71
Reagents and conditions:
(a)
30% NaOMe in MeOH, MeOH, 40 °C, 1 h; (b) aminoacetaldehyde dimethylacetal,
AcOH, MeOH, reflux temperature, 30 min; (c) 6 M HCl, reflux temperature,
18 h; (d) MeI, NaH, dry DMF, rt, 2 h; and (e) 4-morpholinoaniline,
1.25 M HCl in EtOH, n-BuOH, 180 °C, 16 h.
Synthesis of Imidazol-2-yl Pyridine
Derivative 71
Reagents and conditions:
(a)
30% NaOMe in MeOH, MeOH, 40 °C, 1 h; (b) aminoacetaldehydedimethylacetal,
AcOH, MeOH, reflux temperature, 30 min; (c) 6 M HCl, reflux temperature,
18 h; (d) MeI, NaH, dry DMF, rt, 2 h; and (e) 4-morpholinoaniline,
1.25 M HCl in EtOH, n-BuOH, 180 °C, 16 h.For the synthesis of compounds 75 and 78, presenting a methylaminothiazole central core, an approach
related
to Hantzsch thiazole synthesis[32] was employed
(Schemes and 9). Thiazole 75 was obtained starting
from 1-(4-fluorophenyl)-2-(2-fluoropyridin-4-yl)ethan-1-one (72),[23] whereas compound 78 was synthesized starting from 1-pyridinyl-propan-1-one (22). Both ketones 72 and 22 were monohalogenated
at the α-position under acidic conditions and then cyclized
via N-methylthiourea, affording intermediates 74 and 77, respectively. Conclusively, substitution
with 4-morpholinoaniline yielded the desired compounds 75 and 78.
Scheme 8
Synthesis of 2-Methylaminothiazole 75
Reagents and conditions: (a)
Br2, 30% HBr in AcOH, 75 °C, 2 h; (b) N-methylthiourea, EtOH, reflux temperature, 1 h; and (c) 4-morpholinoaniline,
1.25 M HCl in EtOH, n-BuOH, 180 °C, 16 h.
Scheme 9
Synthesis of 2-Methylaminothiazole 78
Reagents and conditions: (a)
Br2, HBr 30% in AcOH, 75 °C, 4 h; (b) N-methylthiourea, EtOH, reflux temperature, 1 h; and (c) 4-morpholinoaniline,
1.25 M HCl in EtOH, n-BuOH, 180 °C, 16 h.
Synthesis of 2-Methylaminothiazole 75
Reagents and conditions: (a)
Br2, 30% HBr in AcOH, 75 °C, 2 h; (b) N-methylthiourea, EtOH, reflux temperature, 1 h; and (c) 4-morpholinoaniline,
1.25 M HCl in EtOH, n-BuOH, 180 °C, 16 h.
Synthesis of 2-Methylaminothiazole 78
Reagents and conditions: (a)
Br2, HBr 30% in AcOH, 75 °C, 4 h; (b) N-methylthiourea, EtOH, reflux temperature, 1 h; and (c) 4-morpholinoaniline,
1.25 M HCl in EtOH, n-BuOH, 180 °C, 16 h.
Biological Evaluation
All synthesized
inhibitors were
evaluated by enzyme-linked immunosorbent assays[33,34] to determine their ability to inhibit JNK3 and p38α MAPK,
and the results are presented in Tables –4 and 6.
Table 1
Core Modifications on 4-F-Phenyl-Substituted
Derivativesa
Data of standard inhibitors SB203580 (p38α MAPK) and SP600125 (JNK3)
in our in-house activity assay are included.
IC50 values are the mean
of three experiments.
n = 16.
n = 20.
Table 4
Effect of Small Alkyl Substituents
in the HR I
IC50 values are the mean
of three experiments.
Percent
inhibition at indicated
concentration.
Data of standard inhibitors SB203580 (p38α MAPK) and SP600125 (JNK3)
in our in-house activity assay are included.IC50 values are the mean
of three experiments.n = 16.n = 20.IC50 values are the mean
of three experiments.Percent
inhibition at indicated
concentration.IC50 values are the mean
of three experiments.Percent
inhibition at indicated
concentration.IC50 values are the mean
of three experiments.Percent
inhibition at indicated
concentration.The free
terminal aniline moiety of compound 1a is
considered to be potentially responsible for aggregation and therefore
might result in assay interference, as also pointed out by analysis
through the ZINC 15 pattern tool.[35] For
this reason, the p-phenylendiamine moiety at the
pyridine-C2 position of compound 1a was modified in a
4-morpholinoaniline group, which has already been reported as a beneficial
substituent in this position.[36] Resulting
compound 5 (Table ) displayed extremely close inhibition values to its analogoue 1a (1a, IC50(JNK3) = 24 nM; IC50(p38α MAPK) = 17 nM) and this moiety was, therefore,
maintained constant during the investigation of other positions of
the scaffold.The first attempt, which was carried out to shift
the preference
of compound 5 toward the JNK3, consisted of modifying
the central imidazole core together with acting on the substitution
at the imidazole-C2 position (Table ). Transformation of the methylsulfanyl group at the
imidazole-C2 position into an ethyl group or removal of the same group,
resulting in compounds 13 and 14, respectively,
did not seem to affect the inhibitory activity on the two enzymes.
Replacement of the imidazole core with an imidazol-2-one ring instead
caused a decrease in the JNK3 inhibitory activity while leaving the
inhibition of p38α MAPK unchanged (8: IC50(JNK3) = 142 nM; IC50(p38α MAPK) = 34 nM). The position
of the two nitrogen atoms at the central imidazole core seems to be
essential for the inhibition of both enzymes, as the different arrangements
of substituents around the five-membered ring of 1,5-disubstitutedimidazole 66 resulted in a drop in activity on both target
kinases. On the other hand, exchange of 2-sulfanylimidazole with 2-methylaminothiazole
(75) yielded an increase in inhibitory activity of 2.5-
and 8-fold for JNK3 and p38α MAPK, respectively.To assess
the effect of the substituent located in the hydrophobic
region (HR) I, the 4-fluorophenyl group was replaced by different
aromatic, alkyl, and cycloalkyl moieties (Table ). In terms of both ligand efficiency (LE)
as well as lipophilic LE (LLE), the 4,5-disubstituted derivative 14 is the most efficient one out of the series of Table and serves, therefore,
as the optimal starting point for these modifications. Moreover, this
scaffold presents a substantially equal activity compared to its S-methylated
analogue 5, along with a convenient synthetic strategy,
facilitating the preparation of a broad range of derivatives.Most of the 4,5-disubstituted pyridinylimidazoles having an aromatic
moiety at the imidazole-C4 position (compounds 18a–f) revealed to be potent inhibitors for both enzymes, displaying IC50 values down to the low double digit nanomolar range. In
general, addressing the HR I with a phenyl or monosubstituted phenyl
ring resulted in dual inhibitors displaying a slight preference toward
p38α MAPK over JNK3. This trend is most distinct in the case
of compound 18d having a 3-(trifluoromethyl)phenyl moiety,
which presents a 6-fold higher activity for p38α MAPK than for
JNK3. The only aromatic substituent stepping out of this trend was
the heteroaromatic N-methylpyrazole of compound 18f, producing an overall decrease in activity on both kinases
while conserving a moderate preference toward JNK3 (18f: IC50(JNK3) = 758 nM; IC50(p38α MAPK) = 3259 nM). These findings indicate that substitution on the phenyl
ring is not beneficial when pursuing selectivity on JNK3 and instead
seems to be counterproductive, increasing the activity on p38α
MAPK. The reason behind this lack of selectivity can be intuitively
explained by considering the dimensions of the hydrophobic pocket
known as HR I in the two target kinases. This cleft is wider in the
p38α MAPK than in JNK3 mostly because of a difference in the
“gatekeeper” residue (Thr106 in p38α MAPK vs Met146
in JNK3). However, as already mentioned in some cocrystallization
studies,[37] aromatic moieties can induce
a shift of the flexible side chain of Met146 (JNK3), thus essentially
abolishing the size differences between the two pockets. As a proof
of that, even the bulky 2-naphthyl group of compound 18e seems to be accommodated in the “reshaped” hydrophobic
pocket of JNK3, therefore resulting in a high inhibitory potency.
Moreover, attempts of substituting the ortho and meta positions of
the phenyl ring, seeking for additional interactions, did not succeed
and produced negative outcomes instead (compounds 18b–d).The replacement of the aromatic ring at the imidazole-C4
position
by cycloalkyl moieties resulted in a dramatic decrease in activity
for both enzymes, with IC50 values in the low micromolar
range. The only exception was the cyclohexyl derivative 18g that was able to interact with p38α MAPK with an IC50 value of 726 nM, 2-fold more potent than on JNK3. The inhibitory
effect of compounds 18g–j on p38α MAPK,
decreasing alongside the reduction of the ring size, is symptomatic
of a gradually diminished capability of the cyclic group to occupy
the spacious cavity of the enzyme. On the other side, JNK3 activity
of derivatives 18g–i, bearing a four- to six-membered
ring at the imidazole-C4 position, remained substantially constant,
although significantly decreased compared to the parent compound 14. An analogous scenario occurred in the case of compounds
featuring branched aliphatic groups at the same position. The isopropyl
derivative 18l, analogously to the closely related 18j, showed a significant drop in activity on p38α MAPK,
while conserving an IC50 value on JNK3 in the low micromolar
range. On the other hand, introduction of a tert-butyl
moiety (18k) resulted in a complete loss of activity
on both JNK3 and p38α MAPK. Because of their flexibility and
low electron density, cyclic and branched aliphatic groups are presumably
unable to promote the Met146 shift and therefore cannot fit in the
narrow hydrophobic back pocket of JNK3. A reasonable consequence of
this would therefore consist of the flip of the imidazole ring, directing
the branched or cyclic alkyl moieties away from the hydrophobic back
pocket of the JNK3, thus explaining the similarity of the inhibitory
activity regardless of the substituent size.In agreement with
the trend of the series, methyl- and ethyl-substituted
imidazoles 38 and 39, respectively, displayed
no inhibition of the p38α MAPK (IC50 > 10 μM),
however, preserving activity on the JNK3. In particular, the methyl
derivative 38 represented the sole compound of this series
reaching a submicromolar activity on JNK3 without any remarkable effect
on the p38α MAPK. Moreover, this inhibitor also represents the
most efficient selective inhibitor of this series in terms of LE and
LLE and was therefore chosen as the starting point for further investigations.Once the methyl substituent at the imidazole-C4 position was selected,
our attention was refocused on the central core (Table ). Because of the presence of
the methyl substituent, all derivatives presented in this series lost
their potency on the p38α MAPK, with each one displaying an
IC50 value higher or equal 10 μM. Altering the arrangement
of the substituents around the imidazole ring proved beneficial in
the case of the 1,5-disubstituted imidazole 67, slightly
increasing its potency compared to the precursor 38,
whereas it was deleterious for the 1,2-disubstituted derivative 71. Replacement of the imidazole core with a 2-aminomethyl
thiazole (78) also revealed to be detrimental for the
inhibitory activity. A different approach consisted of the introduction
of an additional substituent on the imidazole-N atom, together with
a reintroduction of the S-methyl group at the C2 position, yielding
the tetrasubstituted imidazole scaffold already reported in potent
dual JNK3/p38α MAPK as well as JNK3selective inhibitors.[16,21] In the case of p38α MAPK, the effect of an additional alkyl
substituent on the imidazole ring has been reported to be strictly
dependent on the position of the substituted N atom. Several examples
have demonstrated alkylation of the imidazole-N atom away from the
pyridine ring to cause a severe reduction of the activity because
of the impossibility to establish a hydrogen bond with the Lys53 of
the p38α MAPK.[38,39] Because the same interaction
has shown to also occur in the binding to JNK3 (Lys93), an analogous
effect was expected on this enzyme as well and was confirmed by the
remarkably reduced JNK3 inhibition by compound 50, carrying
a methyl group on the distal imidazole N atom. On the other hand,
becauseseveral tetrasubstituted JNK3/p38α MAPK inhibitors have
been reported with an alkyl substituent on the imidazole N adjacent
to the pyridine ring, we assumed this modification to be suitable
with our 4-methyl substituted scaffold as well. However, derivatives 57 and 58, featuring a methyl and an ethyl substituent
on the N atom proximal to pyridine, respectively, unexpectedly presented
an even lower potency on JNK3 than the supposedly “wrong”
regioisomer 50. The drop in activity appeared to increase
with the size of the alkyl substituent, as N-cyclopropyl substituted 59 was almost 3-fold less active compared to its N-methyl
analogue 57. This outcome suggests that despite not hampering
the formation of a H bond with the Lys93, alkyl substituents at the
imidazole N atom proximal to pyridine reduce the tightness of the
binding to the JNK3 active site.
Table 3
Modification of the Core on Methyl-Substituted
Derivatives
IC50 values are the mean
of three experiments.
Percent
inhibition at indicated
concentration.
To complete this series, starting
from 4,5-disubstituted imidazole 38, the original S-methyl
group or a free amino substituent
was introduced at the imidazole-C2 position, affording compounds 44 and 47, respectively. Although the 2-amino
imidazole derivative showed a drop in activity compared to the parent
compound 38, reintroduction of the S-methyl group at
the imidazole-C2 position surprisingly produced a 2-fold increase
in the inhibitory potency on JNK3 (44: IC50(JNK3) = 363 nM; IC50(p38α MAPK) > 10 μM).
This outcome prompted us to reconsider our previous assumption regarding
the role of the 2-methylsulfanyl moiety. Although the S-methyl group
exerts no influence on the inhibitory activity when the 4-fluorophenyl
moiety is installed at the imidazole-C4 position, it has a significant
impact in the case of 4-methyl imidazole derivatives.In a closer
evaluation concerning the influence of the alkyl chain
in position 4 of the imidazole core combined with the 2-methylsulfanyl
moiety in C2 position, the methyl group (44) emerged
once more as the substituent presenting the optimal length to target
the JNK3 HR I, when compared to the 4-unsubstituted and to the 4-ethyl
derivatives 43 and 45, respectively (Table ). Comparison of imidazoles 5 (Table )
and 44 (Table ) reveals the replacement of the 4-fluorophenyl ring at the
imidazole-C4(5) position by a smaller methyl group to result in a
1 order of magnitude loss in JNK3 inhibition and in a complete loss
of p38α MAPK inhibitory activity.To elucidate the binding
mode of the 4-methyl-substituted-5-(pyridine-4-yl)imidazole
derivatives, as well as to gain insight into the role of the 2-methylsulfanyl
group, crystal structures of JNK3 in complex with compounds 38 and 44 were determined (Figure ).
Figure 2
Crystal structures of JNK3 in complex with inhibitors 38 (A) and 44 (B) featuring a pyridinylimidazole
scaffold.
Only the JNK3 active site is shown. The protein backbone is displayed
in gray. The compounds, the side chain of gatekeeper Met146, and a
part of the Gly-rich loop are highlighted in stick display. Active
site residues with common orientations and interactions are shown
in light blue, whereas residues that differ between both complexes
are highlighted in the same color as the respective inhibitor. Side
chains for which multiple orientations are observed (Asn194 in complex
with 38 and Asn152 in complex with 44) are
shown in both orientations. Water molecules are represented as red
spheres and hydrogen bonds are shown as black dashed lines.
Crystal structures of JNK3 in complex with inhibitors 38 (A) and 44 (B) featuring a pyridinylimidazole
scaffold.
Only the JNK3 active site is shown. The protein backbone is displayed
in gray. The compounds, the side chain of gatekeeperMet146, and a
part of the Gly-rich loop are highlighted in stick display. Active
site residues with common orientations and interactions are shown
in light blue, whereas residues that differ between both complexes
are highlighted in the same color as the respective inhibitor. Side
chains for which multiple orientations are observed (Asn194 in complex
with 38 and Asn152 in complex with 44) are
shown in both orientations. Water molecules are represented as red
spheres and hydrogen bonds are shown as black dashed lines.The structures revealed a similar
binding mode of the inhibitors
within the adenosine 5′-triphosphate (ATP) pocket of JNK3 (Figure ). As expected, both
scaffolds interacted with the hinge region of the kinase via two hydrogen
bonds involving the main chain carbonyl and backbone amine groups
of Met149 and mimicking the interactions of the enzyme with ATP[40] as well as with its nonhydrolyzable analogue
β,γ-methyleneadenosine-5′-triphosphate (AMP–PCP,
Figure S3, Supporting Information). In
both structures, the imidazole-N atom distal from the pyridine ring
is part of a network of water-mediated hydrogen bonds, involving the
side chain of Lys93 and the main chain of Leu206. Further water-mediated
hydrogen bonds in the JNK3-38 crystal structure (Figure A) include the side
chain of Asn194, whereas in the JNK3-44 structure (Figure B), the backbone
of Gly76 and the side chain of Asp207 are involved. The structure
of JNK3 in complex with inhibitor 38 also showed that
the imidazole-N atom proximal to the pyridine ring participates in
a water-mediated hydrogen bond with the Asn152 side chain and the
same interaction seems to be present in the JNK3-44,
thus explaining the detrimental effect produced by the substitution
of this position (compounds 57–59). Multiple hydrophobic
interactions comprising the gatekeeperMet146 and the side chains
of Ile70, Val78, Val196, and Leu206 were also observed. These interactions
have been previously described by Scapin et al.[37] and confer JNK3selectivity as they cannot be formed in
the larger binding pocket of p38α MAPK. The methyl group present
in both inhibitors was oriented toward the HR I, which resulted in
an identical orientation of the side chain of the gatekeeper residue
Met146. The 4-morpholinoaniline moiety, which occupied the solvent-exposed
HR II, exhibited higher flexibility and no direct interactions with
JNK3, that is, this moiety likely contributes barely or not at all
to the binding.A major structural difference between the two
complex structures
was observed for the Gly-rich loop. In the JNK3-38 complex
structure, no electron density for residues Gly71–Gly76 was
visible because of high local flexibility, a phenomenon also encountered
in other JNK3 crystal structures.[41−43] In the JNK3-44 complex, however, the electron density for this loop was clearly
defined, hinting to a structural stabilization of this region upon
interaction with the 2-methylsulfanyl moiety in compound 44.A superposition of our inhibitor complex structures with
crystal
structures of JNK3 bound to AMP–PCP and the dual JNK3/p38α
MAPK inhibitor by Scapin et al.[37] (PDB
code: 1PMN)
yielded insights into the structural basis for the observed selectivity
of compounds 38 and 44 (Figure ).
Figure 3
Comparison of the gatekeeper
Met146 orientation and the Gly-rich
loop positioning upon JNK3 inhibitor binding with other ligand-bound
JNK3 structures. Overlay of the JNK3-44 complex structure
(light green), the JNK3-38 complex structure (light red),
the AMP–PCP-bound JNK3 structure (light orange), and the 1PMN structure reported
by Scapin, et al.[37] (blue). The superposition
was performed using the “align” function in PyMOL. The
side chains of the gatekeeper Met146 and the Gly-rich loop are highlighted.
Only compounds 38 and 44 are shown for the
sake of clarity.
Comparison of the gatekeeperMet146 orientation and the Gly-rich
loop positioning upon JNK3 inhibitor binding with other ligand-bound
JNK3 structures. Overlay of the JNK3-44 complex structure
(light green), the JNK3-38 complex structure (light red),
the AMP–PCP-bound JNK3 structure (light orange), and the 1PMN structure reported
by Scapin, et al.[37] (blue). The superposition
was performed using the “align” function in PyMOL. The
side chains of the gatekeeperMet146 and the Gly-rich loop are highlighted.
Only compounds 38 and 44 are shown for the
sake of clarity.As can be seen from this
structural comparison, no movement of
the gatekeeperMet146 side chain is induced by compounds 38 and 44 when compared to the AMP–PCP complex,
contrary to the dual kinase inhibitor studied by Scapin et al. In
the latter crystal structure, an induced fit of side chain 146 occurred
to accommodate the dichlorophenyl moiety of the dual kinase inhibitor.
Conversely, it appears that the methyl substituent of compounds 38 and 44 was unable to occupy the wider HR I
of the p38α MAPK, while possessing the optimal length to target
the respective region of JNK3. Therefore, this moiety determined the
selectivity achieved over p38α MAPK, demonstrated by the activities
of compounds listed in Tables −4. In the case of AMP–PCP
and compound 44, another result of the interaction is
a downward positioning of the flexible Gly-rich loop. A similar compression
of the binding pocket caused by a repositioning of the Gly-rich loop
was reported for a JNK3 complex crystal structure by Kamenecka et
al.[44] and might be a result of hydrophobic
interactions and water-mediated hydrogen bonds provided by inhibitor 44, which stabilized this otherwise flexible section. Overall,
as a result of inhibitor binding, the JNK3ATP binding pocket in our
crystal structures appears somewhat narrower in comparison to the
p38α MAPK binding site (where the gatekeeper is Thr106), an
effect that is less prominent for the dual kinase inhibitor (Figure ) and probably responsible
for the selectivity of compounds 38 and 44. With respect to the 2-fold increase in the inhibitory potency on
JNK3 of compound 44 over its analogue 38, the influence of the S-methyl group on the positioning of the Gly-rich
loop is the most likely structural reason for the significant gain
in affinity.An additional characterization of the two compounds 38 and 44 included the determination of the protein
melting
temperature (Tm) in the presence and absence
of inhibitors by nano differential scanning fluorimetry (nanoDSF).
This methodology consists of assessing the influence of the binding
event on the stability of the target protein and is carried out by
monitoring temperature-dependent changes in the intrinsic protein
fluorescence as a consequence of unfolding. The corresponding curves
(Figure S2, Supporting Information) exhibited
a significant increase in stability of JNK3 upon inhibitor binding,
as can be seen from the associated Tm values
(Table ). The Tm value of JNK3 alone was determined to be 46.3
°C and increased to 53.8 and 54.8 °C in the presence of
compounds 38 and 44, respectively, which
correlates well with the results concerning the inhibitory activity
and stability of the Gly-rich loop.
Table 5
Determined Melting
Temperatures (Tm) for JNK3 Alone and in
Complex with Inhibitors 38 and 44
sample
Tm (°C)a
JNK3
46.28 ± 0.58
JNK3-38
53.87 ± 0.04
JNK3-44
54.83 ± 0.04
Data represent mean value ±
SD of a single experiment performed in triplicate. nanoDSF measurements
(Figure S2, Supporting Information) were
conducted using Prometheus NT.48 (NanoTemper Technologies, Munich).
Data represent mean value ±
SD of a single experiment performed in triplicate. nanoDSF measurements
(Figure S2, Supporting Information) were
conducted using Prometheus NT.48 (NanoTemper Technologies, Munich).A further approach in the pursuit
of a tighter binding with the
JNK3 consisted of modifying the amino moiety at the pyridine-C2 position
(Table ).IC50 values are the mean
of three experiments.Percent
inhibition at indicated
concentration.According
to the ZINC patterns tool,
compound 48m represents a potential pan-assay interference
compound. However, this compound was synthesized as the intermediate
for the preparation of inhibitors 48n–q. To estimate
the impact of the amide moiety present in compounds 48n–q on the inhibition of the two kinases, the activities of 48m are listed in this table.An initial attempt was carried out by introducing α-methyl(phenyl)alkylamino
moieties (compounds 48a–b) as well as cycloalkylamino
groups (compounds 48c–e). The former moieties
have been reported in potent p38α MAPK inhibitors, for example, LN950 (Figure ) and ML3403,[45] and were
thus introduced to evaluate their effect on JNK3 inhibitory potency.
In detail, these substituents were hypothesized to yield an increase
in the JNK3 inhibitory activity while conserving selectivity over
the p38α MAPK because of the combination with the 4-methyl substituent
on the imidazole ring. However, the 3-methyl-2-butylamino group (48a) resulted in a loss of activity compared to the 4-morpholinoaniline
precursor 44, although maintaining some selectivity over
the p38α MAPK. Substitution with the α-methylbenzylamine,
giving rise to compound 48b, was instead counterproductive
as it not only caused a tremendous drop in JNK3 inhibition but also
a recovery of the activity on the p38α MAPK (48b: IC50(JNK3) = 7610 nM; IC50(p38α MAPK) = 3460 nM). On the other hand, although not reaching the potency
of the parent compound 44, the JNK3 inhibitory activity
of compounds bearing cycloalkylamino moieties at the pyridine-C2 position
(48c–e) increased alongside the size of the aliphatic
ring, a trend suggesting the importance of hydrophobic interactions
in this area of the molecule. Nevertheless, replacement of the cyclohexyl
ring of 48e with the similar tetrahydropyranyl group
(48f) yielded, unexpectedly, a remarkable loss of activity
on JNK3.A possible strategy to gain activity and selectivity
on JNK3 would
consist of targeting the side chain of Gln155 as this residue is replaced
by a shorter Asn in the p38α MAPK.[46] As suggested by the structure of the JNK3–44 complex (Figure ), this amino acidic residue is located about 4 Å away from
the 4-morpholinoaniline-N atom but cannot be reached because of the
rigidity of this substituent. Moreover, the 4-morpholinoanilino moiety
is only able to accept a hydrogen bond, whereas the Gln residue has
the potential to act as both acceptor and donor of hydrogen bonds.
For this reason, trans-4-aminocyclohexanol and trans-1,4-diaminocyclohexyl moieties were selected for compounds 48g and 48h, respectively, because of a higher
flexibility and their additional capability to donate hydrogen bond
interactions. In particular, the former moiety is also present in
the structure of clinical candidate CC-930, wherein it is reported
to interact with the aforementioned Gln155,[47] and included in potent p38α MAPK inhibitors.[45] Unfortunately, despite preserving the selectivity over
the p38α MAPK, none of the two inhibitors 48g and 48h succeeded in overcoming the activity of the parent compound 44 on the JNK3, with the latter displaying a 3-fold drop in
potency. This observation suggests an inability of the introduced
moiety to form the desired interaction with the Gln155 side chain
or this interaction being compensated by other factors. Additionally,
it underlines the necessity of the aromatic moiety at the pyridine-C2
amino function for the binding to the JNK3. The significantly lower
activity of compound 48h could also derive from the not
tolerated protonation of its terminal amino functionality. With the
aim to reach the Gln155 side chain by the introduction of an additional
hydrogen bond acceptor, a series of amides of compound 48h and of its aromatic counterpart 48m was synthesized.
This approach also permits to seek additional interactions with the
enzyme HR II. Unfortunately, in neither of the two series, the introduction
of amide moieties permitted to gain an inhibitory activity comparable
with the precursor 44. In the series featuring a cycloaliphatic
amine (48i–l), only the small acetamide derivative 48i exhibited an almost similar activity to the precursor,
whereas bulkier alkyl and aromatic residues displayed a 2- to 3-fold
decrease in potency. In an analogous fashion, when considering the
series derived from the aromatic intermediate 48m, compounds
bearing a tert-butyl or a cyclohexyl amide (48p and 48q, respectively) showed a significant
drop in inhibitory activity, with IC50 values in the micromolar
range. On the other hand, both inhibitors carrying an acetamido or
benzamido moiety (48n and 48o, respectively)
were still able to inhibit the JNK3 with a potency akin to the free
amine derivative 48m. The comparison of the two amideseries also supports the theory of a higher suitability of aromatic
substituents at the pyridine-C2 amino position when targeting the
JNK3.Compound 44 resulted as the best inhibitor
of the
synthesized series and was, therefore, further investigated to achieve
a comprehensive characterization. At a first instance, to evaluate
the intra-JNK selectivity, compound 44 was tested on
the three JNK isoforms (Table ). As expected, compound 44 inhibited the three
isoforms with a similar potency but showed a moderate preference for
JNK1 and JNK3 over JNK2.
Table 7
Inhibition Data of
Compound 44 on the Three JNK Isoforms
IC50 [nM]a
JNK1
JNK2
JNK3
119
468
184
Compound 44 was tested
by Reaction Biology corporation (Malvern, PA, USA) using a radiometric
assay.
Compound 44 was tested
by Reaction Biology corporation (Malvern, PA, USA) using a radiometric
assay.Moreover, inhibitor 44 was further screened against
a panel of 45 diverse kinases to achieve a preliminary evaluation
of its selectivity within the kinome. Out of the kinase panel, 10
kinases (including JNK1) were inhibited more than 50% at a testing
concentration of 10 μM (Table S3, Supporting Information).Additional studies were aimed at evaluating
the inhibition of the
human-ether-à-go-go related gene (hERG) potassium channels
as well as liver cytochromes P450 (CYP450) to highlight potential
liabilities of the synthesized scaffold. As displayed in Table , compound 44 showed a reduced interference with the hERG channels (IC50 > 10 μM).
Table 8
Inhibitory Activity
of Compound 44 on hERG Channels and on the Most Relevant
CYP Isoforms
CYP450
inhibition [% inhibition at 10 μM]
hERG inhibition [% inhibition at 10 μM]
1A9
2C9
2C19
2D6
3A4
38.8
51.5
53.9
35.6
19.0
75.1
Regarding interaction with hepatic
enzymes, compound 44 displayed low to moderate inhibition
of four of the five tested
isoenzymes, this representing a significantly cleaner profile in comparison
with previously reported inhibitors of this class.[48] However, the elevated blockage of the most abundant CYP450
isoform 3A4 still constitutes a serious limit, which needs to be solved
by subsequent optimization strategies.Finally, additional tests
were performed to assess the metabolic
stability of methyl-substituted pyridinylimidazole 44 upon incubation with human liver microsomes. One of the most serious
limitations of previously reported 2-alkylsulfanylimidazoles is their
severe metabolization consisting of oxidation of the thioether function
to the corresponding sulfoxide.[48] Nevertheless,
in vitro assays performed on compound 44 demonstrated
a substantial metabolic stability, as approximately 80% of the unmodified
compound was still present after 4 h incubation (Figure S6, Supporting Information). The major metabolite
formed still appears to be represented by the sulfoxide derivative
(8.49%), although modifications at the 4-morpholinoaniline substituent
might also be present.
Conclusions
Optimization of 4-(4-fluorophenyl)-5-(pyridin-4-yl)imidazole-based
p38α MAPK inhibitors by modification of the five-membered heterocyclic
core, the aryl moiety at the imidazole-C4 position, and the pyridine-C2
amino function resulted in a novel series of JNK3 inhibitors exhibiting
high selectivity over the closely related p38α MAPK. Biological
evaluation of the different pyridinyl-substituted five-membered rings
provided valuable insights into the structure activity relationship
of this scaffold with respect to JNK3 and p38α MAPK inhibitory
potencies. By addressing the HR I with a small methyl group, a significant
selectivity toward JNK3 was achieved. This feature is not yet reported
for this class of compounds, which have been generally described as
p38α MAPK inhibitors. The binding mode at the ATP binding site
of the enzyme for this class of compounds was confirmed by X-ray structures
of JNK3 crystals incubated with imidazoles 38 and 44. The most potent inhibitor 4-(4-methyl-2-(methylthio)-1H-imidazol-5-yl)-N-(4-morpholinophenyl)pyridin-2-amine
(44) inhibits the JNK3 in the low triple digit nanomolar
range, is metabolically stable, and displays a slight selectivity
over the JNK2 isoform. Further characterization of this inhibitor
highlighted reduced interactions with the hERG channel as well with
most of the tested CYP450 isoforms.
Experimental Section
General
All chemicals were purchased from commercial
sources unless otherwise specified and used without further purification.
Thin-layer chromatography (TLC) reaction controls were performed for
all reactions using fluorescent silica gel 60 F254 plates
(Merck) and visualized under natural light and UV illumination at
254 and 366 nm. The purities of all tested compounds were confirmed
to be >95% as determined by reverse-phase high-performance liquid
chromatography (HPLC) using one of the two following methods. In the
case of method 1, the instrument used was a Hewlett Packard HP 1090
Series II LC equipped with a UV diode array detector (DAD) (detection
at 230 and 254 nm). The chromatographic separation was performed on
a Phenomenex Luna 5u C8 column (150 mm × 4.6 mm, 5 μm)
at 35 °C oven temperature. The injection volume was 5 μL
and the flow was 1.5 mL/min using the following gradient: 0.01 M KH2PO4, pH 2.3 (solvent A), MeOH (solvent B), 40%
B to 85% B in 8 min; 85% B for 5 min; 85% to 40% B in 1 min; 40% B
for 2 min; stop time 16 min. In the case of method 2, an Agilent 1100
Series HPLC system was used, equipped with a UV DAD (detection at
218, 254, and 280 nm). The chromatographic separation was performed
on an XBridge C18 column (150 mm × 4.6 mm, 5 μm) and the
oven temperature was set to 30 °C. The injection volume was 10
μL and the flow was 1.5 mL/min using the following gradient:
0.01 M KH2PO4, pH 2.3 (solvent A), MeOH (solvent
B), 45% B to 85% B in 9 min; 85% B for 6 min; stop time 16 min. Flash
column chromatography was performed using an Interchim puriFlash 430
automated flash chromatography system with Davisil LC60A 20–45
μm silica from Grace Davison and Geduran Si60 63–200
μm silica from Merck for the precolumn. Nuclear magnetic resonance
(NMR) data were obtained on a Bruker ARX NMR spectrometer at 250 MHz,
on a Bruker AVANCE III HD NMR spectrometer at 300 MHz, or on a Bruker
AVANCE NMR spectrometer at 400 MHz at ambient temperature. Chemical
shifts are reported in parts per million (ppm) relative to tetramethylsilane.
All spectra were calibrated against the (residual proton) peak of
the deuterated solvent used. Mass spectra were recorded on an Advion
expression S electrospray ionization mass spectrometer (ESI-MS) with
TLC interface.
Experimental Procedures
General Procedure
for the Nucleophilic Aromatic Substitution
with 4-Morpholinoaniline (General Procedure A)
In a pressure
vial, the 2-halide pyridine intermediate (1 equiv) and 4-morpholinoaniline
(1.5 equiv) were suspended in n-butanol (3 mL) and
1.25 M HCl in EtOH (1 equiv) was added. After tightly closing the
vial, the reaction mixture was heated in a heating block at 180 °C
and stirred for 18 h. After removing the solvent at reduced pressure,
the residue was purified by flash column chromatography.
General Procedure
for the Synthesis of Compounds 15a–l (General
Procedure B)
In a three-neck round-bottom flask
under anhydrous conditions, 2-chloro-4-methylpyridine (9) (1 equiv) and the appropriate ethyl ester (1 equiv) were dissolved
in dry tetrahydrofuran (THF) (2 mL). After cooling the reaction mixture
to 0 °C, 2 M sodium bis(trimethylsilyl)amide (NaHDMS) in dry
THF (2.2 equiv) was added dropwise and the mixture was stirred at
0 °C for 1.5–5 h. After adding H2O, the aqueous
phase was extracted three times with dichloromethane (DCM) or EtOAc
and washed with NaCl saturated solution. The combined organic layers
were dried over anhydrous Na2SO4 and the solvent
was evaporated at reduced pressure. The residue was finally purified
by flash column chromatography.
General Procedure for the
Synthesis of Compounds 16a–l (General Procedure
C)
Ethan-1-one intermediates 15a–l (1
equiv) and SeO2 (1.1 equiv) were suspended in 5–10
mL of glacial AcOH and the reaction mixture was stirred at 65 °C
for 2–3 h. After cooling to room temperature (rt), the formed
solid residue of Se was removed by filtration and the filtrate was
diluted with EtOAc and then washed with saturated NaHCO3 solution four times. Finally, the organic phase was washed with
saturated NaCl solution, dried over anhydrous Na2SO4, and concentrated at reduced pressure. The residue was purified
by flash column chromatography.
General Procedure for the
Synthesis of Compounds 17a–l (General Procedure
D)
In a pressure vial, ethane-1,2-dione
derivatives 16a–l (1 equiv) and NH4OAc (10 equiv) were suspended in 3 mL of glacial AcOH and after that
a 37% aqueous solution of formaldehyde (1 equiv) was added. The reaction
vessel was heated in a CEM microwave reactor at 180 °C, with
an initial power of 200 W, for 2–5 min. The mixture was added
dropwise to NH4OH concentrated solution at 0 °C. The
suspension obtained was extracted three times with EtOAc and the combined
organic layers were dried over anhydrous Na2SO4 and concentrated at reduced pressure. The residue was purified by
flash column chromatography.
General Procedure for the
Synthesis of Compounds 48a–h (General Procedure
E)
In a pressure vial, 2-chloro-4-(4-methyl-2-(methylthio)-1H-imidazol-5-yl)pyridine (41) was suspended
in ≈2 mL of cycloalkylamine (in the case of solid amine, 20
equiv of amine was added and the mixture was suspended in ≈2
mL of n-butanol). The closed vial was then heated
at 180 °C and stirred for 48–120 h. The reaction mixture
was poured in H2O and the aqueous layer was extracted three
times with EtOAc. The combined organic layers were dried over anhydrous
Na2SO4 and concentrated at reduced pressure.
The residue was finally purified by flash column chromatography.
General Procedure for the Synthesis of Compounds 48i–l and 48n–q (General Procedure F)
Under
an argon atmosphere, trans-N1-(4-(4-methyl-2-(methylthio)-1H-imidazol-5-yl)pyridin-2-yl)cyclohexane-1,4-diamine (48h) or N1-(4-(4-methyl-2-(methylthio)-1H-imidazol-5-yl)pyridin-2-yl)benzene-1,4-diamine (48m) was dissolved in 1.5 mL of dry pyridine and after that
the appropriate acid chloride or anhydride was added and the reaction
mixture was stirred at rt for 16 h. The reaction mixture was poured
in H2O and the aqueous layer was extracted three times
with EtOAc. The combined organic layers were dried over anhydrous
Na2SO4 and concentrated at reduced pressure.
The residue was finally purified by flash column chromatography.
To a solution of 10 (250 mg,
1.01 mmol) in MeOH (5 mL), 7 M ammonia in MeOH (2.89 mL, 20.23 mmol)
and propionaldehyde (88.11 mg, 1.52 mmol) were added and the reaction
mixture was heated to reflux temperature and stirred for 4 h. After
cooling down, the solvent was evaporated at reduced pressure and the
residue was purified by flash column chromatography (SiO2, DCM/EtOH 97:03 to 94:06), obtaining 125 mg of the desired product
(43% yield); 1H NMR (300 MHz, DMSO-d6): δ 1.28 (t, J = 7.6 Hz, 3H), 2.64–2.77
(m, 2H), 7.09 (s, 1H), 7.17–7.40 (m, 3H), 7.47–7.56
(m, 2H), 8.06 (d, J = 5.4 Hz, 1H), 12.41 ppm (br
s, 1H); MS-ESI m/z: [M + H]+ calcd for C16H13F2N3, 286.1; found, 286.0; m/z: [M – H]− calcd for C16H13F2N3, 284.1; found, 284.0; HPLC (method
2): tR = 3.680 min.
Compound 34(21) (1.0 g, 4.43 mmol) was suspended in glacial AcOH (10 mL)
and subsequently 30% H2O2 (602.7 mg, 17.72 mmol)
was added dropwise and the reaction mixture was stirred at rt for
15 min. After adding H2O, the pH was adjusted to 8 using
K2CO3 saturated solution and the aqueous phase
was extracted five times with EtOAc. The combined organic layers were
dried over anhydrous Na2SO4 and concentrated
at reduced pressure, affording 230 mg of the product which was used
in the following step without further purification (25% yield); 1H NMR (300 MHz, DMSO-d6): δ
2.47 (s, 3H), 7.62 (dd, J = 5.3, 1.3 Hz, 1H), 7.65
(br s, 1H), 7.69 (s, 1H), 8.33 ppm (d, J = 5.2 Hz,
1H); 13C NMR (101 MHz, DMSO-d6): δ 11.7, 118.8, 119.1, 127.9, 130.5, 134.9, 145.9, 149.8,
150.8 ppm; MS-ESI m/z: [M + H]+ calcd for C9H8ClN3, 194.0;
found, 194.0; m/z: [M – H]− calcd for C9H8ClN3, 192.0; found, 191.8; HPLC (method 2): tR = 1.375 min.
Compound 35 (400 mg, 1.67
mmol) (for the synthesis of compound 35 see Supporting Information) was suspended in glacial
AcOH (10 mL) and subsequently 30% H2O2 (227.2
mg, 6.68 mmol) was added dropwise and the reaction mixture was stirred
at rt for 40 min. The reaction mixture was concentrated at reduced
pressure and after that 20 mL of K2CO3 saturated
solution was added. The aqueous layer was extracted five times with
EtOAc and the combined organic layers were dried over anhydrous Na2SO4 and concentrated at reduced pressure, affording
230 mg of the product which was used in the following step without
further purification (71% yield); 1H NMR (300 MHz, DMSO-d6): δ 1.22 (t, J = 7.4
Hz, 3H), 2.85 (q, J = 7.4 Hz, 2H), 7.58 (d, J = 5.0 Hz, 1H), 7.62 (s, 1H), 7.70 (s, 1H), 8.33 ppm (d, J = 5.1 Hz, 1H); 13C NMR (101 MHz, DMSO-d6): δ 13.4, 18.8, 119.1, 119.4, 129.7,
133.8, 135.2, 145.8, 149.9, 150.8 ppm; MS-ESI m/z: [M + H]+ calcd for C10H10ClN3, 208.0; found, 208.1; m/z: [M – H]− calcd for C10H10ClN3, 206.0; found, 205.9; HPLC (method
2): tR = 1.653 min.
Under an argon atmosphere, compound 33 (500 mg, 2.36 mmol) (for the synthesis of compound 33 see Supporting Information)
and t-BuONa (454 mg, 4.72 mmol) were dissolved in
dry MeOH (20 mL) and after cooling the reaction mixture to 0 °C,
methyl iodide (147.5 μL, 2.36 mmol) was added and the reaction
mixture was stirred at 0 °C for 30 min. The reaction mixture
was then heated to 55 °C and stirred for 3 h. After cooling to
rt, the solvent was evaporated at reduced pressure and H2O was added. The aqueous phase was then extracted two times with
EtOAc and the combined organic layers were dried over anhydrous Na2SO4 and concentrated at reduced pressure. The residue
was finally purified by flash column chromatography (SiO2, DCM/EtOH 100:0 to 90:10) giving 396 mg of the desired compound
(74% yield); 1H NMR (400 MHz, DMSO-d6): δ 2.59 (s, 3H), 7.64–7.72 (m, 1H), 7.73–7.79
(m, 1H), 8.03 (s, 1H), 8.31 (dd, J = 5.3, 1.8 Hz,
1H), 12.70 ppm (br s, 1H); MS-ESI m/z: [M + H]+ calcd for C9H8ClN3S, 226.0; found, 225.9; m/z: [M – H]− calcd for C9H8ClN3S, 224.0; found, 223.9; HPLC (method 1): tR = 4.096 min.
Cyanamide (652 mg, 15.52 mmol) was dissolved
in EtOH (30 mL) and after heating at reflux temperature, compound 31 was added portionwise over 1 h and the mixture was stirred
at the same temperature further for 3 h. After cooling down, the solvent
was evaporated at reduced pressure and the residue was purified by
flash column chromatography (SiO2, DCM/EtOH/Et3N 95:05:0 to 80:18:2), obtaining 900 mg of the desired product (95%
yield); 1H NMR (300 MHz, DMSO-d6): δ 2.37 (s, 3H), 7.53 (d, J = 4.9 Hz, 1H),
7.62 (br s, 3H), 8.42 (d, J = 5.1 Hz, 1H), 12.86
ppm (br s, 1H); 13C NMR (75 MHz, DMSO-d6): δ 10.7, 117.6, 118.9, 119.4, 124.4, 139.2, 147.0,
150.3, 151.1 ppm; MS-ESI m/z: [M
+ H]+ calcd for C9H9ClN4, 209.0; found, 208.9; m/z: [M
– H]− calcd for C9H9ClN4, 207.0; found, 206.9; HPLC (method 2): tR = 1.524 min.
Under an argon atmosphere, compound 34 (250 mg, 1.11 mmol) and t-BuONa (213 mg,
2.22 mmol) were dissolved in dry MeOH (10 mL), and after cooling the
reaction mixture to 0 °C, methyl iodide (205 μL, 3.32 mmol)
was added and the reaction mixture was let to heat to rt. The reaction
mixture was then heated to 80 °C and stirred for 3 h. After cooling
to rt, the solvent was evaporated at reduced pressure and H2O was added. The aqueous phase was then extracted two times with
EtOAc and the combined organic layers were dried over anhydrous Na2SO4 and concentrated at reduced pressure. The residue
was finally purified by flash column chromatography (SiO2, DCM/EtOH 100:0 to 90:10) giving 110 mg of the desired compound
(39% yield); 1H NMR (400 MHz, DMSO-d6): δ 2.44 (s, 3H), 2.54–2.62 (m, 3H), 3.34 (s,
3H), 7.47–7.77 (m, 2H), 8.34 ppm (d, J = 4.5
Hz, 1H); 13C NMR (101 MHz, DMSO-d6): δ 10.5, 15.4, 30.6, 119.1, 119.4, 130.5, 132.4, 142.3,
145.6, 149.8, 150.8 ppm; HPLC (method 1): tR = 4.812 min.
Under an argon atmosphere, tris(dibenzylidenaceton)dipalladium(0)
(Pd2(dba)3) (17.5 mg, 0.02 mmol) and 9,9-dimethyl-4,5-bis(diphenylphosphino)xanten
(Xantphos) (22.1 mg, 0.04 mmol) were dissolved in dry 1,4-dioxane
(5 mL) and stirred for 10 min. After that compound 49 (50 mg, 0.21 mmol), Cs2CO3 (138.1 mg, 0.42
mmol), and 4-morpholinoaniline (56.7 mg, 0.32 mmol) were added and
the reaction mixture was heated to 100 °C and stirred for 15
h. After cooling to rt, the reaction mixture was diluted with DCM
and the solid residue was removed by filtration. The filtrate was
then concentrated at reduced pressure and the residue was purified
by flash column chromatography (DCM/EtOH 100:0 to 90:10) giving 61
mg of the desired product (73% yield); 1H NMR (400 MHz,
DMSO-d6): δ 2.36–2.44 (m,
3H), 2.54 (m, 3H), 2.93–3.05 (m, 4H), 3.45–3.55 (m,
3H), 3.67–3.76 (m, 4H), 6.79–6.96 (m, 3H), 7.07 (s,
1H), 7.54 (d, J = 7.3 Hz, 2H), 8.00–8.09 (m,
1H), 8.76 ppm (br s, 1H); 13C NMR (101 MHz, DMSO-d6): δ 10.5, 15.8, 30.5, 49.5, 66.2, 106.0,
111.0, 115.9, 119.5, 128.4, 134.4, 134.6, 140.9, 142.9, 145.4, 147.2,
156.7 ppm; MS–FAB m/z: [M]
calcd for C21H25N5OS, 395.2; found,
395.3; HPLC (method 1): tR = 3.156 min
(98.7%).
In a pressure
vial,
compound 31 (200 mg, 0.77 mmol) (for the synthesis of
compound 31 see Supporting Information) and methyl isothiocyanate (284 mg, 3.88 mmol) were suspended in
triethylamine (2 mL), and after closing the vial tightly, the reaction
mixture was stirred at 60 °C for 16 h. The excess of triethylamine
was evaporated at reduced pressure and the residue was suspended in
glacial AcOH and stirred at 80 °C for 1.5 h. The reaction mixture
was concentrated at reduced pressure and after that NaHCO3 saturated solution (20 mL) was added and the aqueous phase was extracted
four times with EtOAc. The combined organic layers were washed with
H2O and NaCl saturated solution, dried over anhydrous Na2SO4, and concentrated at reduced pressure. Finally,
the residue was purified by flash column chromatography (SiO2, DCM/EtOH 100:0 to 95:05), affording 110 mg of the desired product
(60% yield); 1H NMR (300 MHz, DMSO-d6): δ 2.11 (s, 3H), 3.45 (s, 3H), 7.47 (dd, J = 5.2, 1.4 Hz, 1H), 7.55–7.63 (m, 1H), 8.47 (d, J = 5.1 Hz, 1H), 12.51 ppm (br s, 1H); 13C NMR (75 MHz,
DMSO-d6): δ 9.6, 32.4, 122.7, 122.9,
123.3, 124.4, 139.5, 150.2, 150.9, 161.8 ppm; MS-ESI m/z: [M – H]− calcd for
C10H10ClN3S, 238.0; found, 238.0;
HPLC (method 2): tR = 2.353 min.
In a pressure vial, compound 51 (285 mg, 1.19 mmol) and t-BuONa (114.3 mg, 1.19
mmol) were dissolved in dry MeOH (15 mL), and after cooling the reaction
mixture to 0 °C, methyl iodide (217 μL, 3.48 mmol) was
added. The vial was tightly closed and the mixture was stirred at
50 °C for 30 min. After evaporating the solvent at reduced pressure,
H2O was added and the aqueous phase was extracted four
times with EtOAc. The combined organic layers were washed with H2O and NaCl saturated solution, dried over anhydrous Na2SO4, and concentrated at reduced pressure, giving
290 mg of the product which was used in the following step without
further purification (96% yield); 1H NMR (300 MHz, CDCl3): δ 2.28 (s, 3H), 2.66 (s, 3H), 3.52 (s, 3H), 7.12
(dd, J = 5.2, 1.5 Hz, 1H), 7.22–7.24 (m, 1H),
8.44 ppm (d, J = 5.1 Hz, 1H); 13C NMR
(101 MHz, CDCl3): δ 13.7, 15.8, 32.3, 122.0, 123.5,
126.8, 138.5, 141.1, 145.5, 149.9, 152.0 ppm; MS-ESI m/z: [M + H]+ calcd for C11H12ClN3S, 254.0; found, 254.0; HPLC (method
2): tR = 1.720 min.
Under an argon atmosphere, 4-morpholinoaniline
(98.8 mg, 0.55 mmol),
Pd2(dba)3 (16.94 mg, 0.02 mmol), XPhos (17.64
mg, 0.04 mmol), and Cs2CO3 (365 mg, 1.12 mmol)
were placed and after that compound 55 (100 mg, 0.37
mmol) previously dissolved in 5 mL of dry 1,4-dioxane was added and
the reaction mixture was stirred at 100 °C for 18 h. The solvent
was evaporated at reduced pressure and after that NH4Cl
saturated solution was added to the residue and the aqueous phase
was extracted three times with EtOAc. The combined organic layers
were washed with H2O and NaCl saturated solution, dried
over anhydrous Na2SO4, and concentrated at reduced
pressure. Finally, the residue was purified by flash column chromatography
(SiO2, DCM/EtOH 100:0 to 95:05) giving 30 mg of the desired
product (20% yield); 1H NMR (300 MHz, CDCl3):
δ 1.16 (t, J = 7.1 Hz, 3H), 2.20 (s, 3H), 2.63
(s, 3H), 3.05–3.22 (m, 4H), 3.73–3.99 (m, 6H), 6.49–6.64
(m, 2H), 6.81–7.03 (m, 3H), 7.24 (d, J = 8.7
Hz, 2H), 8.19 ppm (d, J = 5.0 Hz, 1H); 13C NMR (101 MHz, CDCl3): δ 13.5, 15.8, 16.1, 39.8,
49.8, 66.9, 107.2, 114.6, 116.8, 124.0, 127.9, 132.3, 137.2, 140.3,
143.0, 148.3, 148.4, 157.7 ppm; MS-ESI m/z: [M + H]+ calcd for C22H27N5OS, 410.2; found, 410.1; m/z: [M – H]− calcd for C22H27N5OS, 408.2; found, 408.1; HPLC (method
2): tR = 2.427 min (98.0%).
2-Bromoisonicotinaldehyde (60, 300 mg, 1.61 mmol), 4-fluoroaniline (179 mg, 1.61 mmol), and AcOH
(160 μL), were dissolved in EtOH and the reaction mixture was
stirred at reflux temperature for 2 h. After cooling to rt, the solvent
was evaporated at reduced pressure and the residue was resuspended
in a mixture 2:1 of MeOH and 1,2-dimethoxyethane (8 mL) and transferred
into a three-neck round-bottom flask under an argon atmosphere. TOSMIC
(471.5 mg, 2.41 mmol) and K2CO3 (445 mg, 3.22
mmol) were added and the mixture was stirred at reflux temperature
for 3 h. The mixture was cooled at rt and the solvent was evaporated
at reduced pressure. The residue was suspended in DCM and the organic
layer was washed three times with H2O and one time with
NaCl saturated solution. The organic phase was dried over anhydrous
Na2SO4 and evaporated at reduced pressure. Finally,
the residue was purified by flash column chromatography (SiO2, DCM to DCM/EtOH 95:05) yielding 360 mg of the desired compound
(70% yield); 1H NMR (400 MHz, DMSO-d6): δ 7.03 (d, J = 4.3 Hz, 1H), 7.30–7.50
(m, 5H), 7.69 (br s, 1H), 8.08 (br s, 1H), 8.24 ppm (d, J = 4.3 Hz, 1H); 13C NMR (101 MHz, DMSO-d6): δ 116.6 (d, J = 22.7 Hz), 120.5,
124.6, 128.3 (d, J = 8.8 Hz), 128.4, 131.9, 132.1
(d, J = 2.9 Hz), 139.5, 141.7, 142.0, 150.3, 161.8
ppm (d, J = 246.6 Hz); MS-ESI m/z: [M + H]+ calcd for C14H9BrFN3, 318.0; found, 317.8; HPLC (method 1): tR = 5.51 min.
Tetrakis(triphenylphosphine)palladium (367
mg, 0,317 mmol) was dissolved in dimethylformamide (DMF) (50 mL) and
after that 5-bromo-1-methyl-1H-imidazole (62) (2.04 g, 12.7 mmol), (2-chloropyridin-4-yl)boronic acid (63) (1.0 g, 6.35 mmol), Cs2CO3 (4.13
g, 12.7 mmol), and H2O (228 mg, 12.7 mmol) were added and
the reaction mixture was stirred at 60 °C for 24 h. The mixture
was poured in H2O and the aqueous phase was extracted five
times with EtOAc. The combined organic layers were washed with NaCl
saturated solution, dried over anhydrous Na2SO4, and concentrated at reduced pressure. The residue obtained was
purified by flash column chromatography (SiO2, DCM/EtOH
95:05 to 90:10), affording 160 mg of the desired product (13% yield); 1H NMR (250 MHz, CDCl3): δ 3.79 (s, 3H), 7.27
(dd, J = 5.2, 1.6 Hz, 2H), 7.34 (br s, 1H), 7.38
(dd, J = 1.6, 0.6 Hz, 1H), 7.60 (br s, 1H), 8.43
ppm (dd, J = 5.2, 0.6 Hz, 1H); MS-ESI m/z: [M + H]+ calcd for C9H8BrN3, 194.0; found, 194.0; HPLC (method 2): tR = 1.162 min.
The
title compound was synthesized according to general procedure A starting
from 65 (140 mg, 0.72 mmol) and 4-morpholinoaniline (192.5
mg, 1.08 mmol). Purification by flash column chromatography (SiO2, DCM/EtOH 100:0 to 90:10) afforded 50 mg of the desired compound
(35% yield); 1H NMR (250 MHz, DMSO-d6): δ 2.97–3.07 (m, 4H), 3.69–3.79 (m,
7H), 6.78–6.85 (m, 2H), 6.89 (m, J = 9.0 Hz,
2H), 7.22 (d, J = 1.2 Hz, 1H), 7.52 (m, J = 9.0 Hz, 2H), 7.76 (br s, 1H), 8.11 (d, J = 5.1
Hz, 1H), 8.85 ppm (br s, 1H); 13C NMR (101 MHz, DMSO-d6): δ 32.9, 49.4, 66.2, 106.9, 111.7,
115.9, 119.9, 128.9, 130.7, 134.0, 137.8, 141.1, 145.8, 147.8, 156.8
ppm; MS-ESI m/z: [M + H]+ calcd for C19H21N5O, 336.2; found,
336.3; m/z: [M – H]− calcd for C19H21N5O, 334.2; found,
334.3; HPLC (method 1): tR = 2.048 min
(100%).
2-Chloro-4-(1H-imidazol-2-yl)pyridine
(69)
To a solution of 2-chloroisonicotinonitrile
(68) (2.0 g, 14.44 mmol) in MeOH (8 mL), a 30% solution
of
NaOMe in MeOH (260 μL, 1.44 mmol) was added and the reaction
mixture was stirred at 40 °C for 1 h. After that both 2,2-dimethoxyethan-1-amine
(1.56 mL, 14.44 mmol) and AcOH (1.56 mL, 27.27 mmol) were added dropwise
and the mixture was stirred at reflux temperature for 30 min. After
cooling to rt, the mixture was diluted with MeOH (8 mL) and then 6
N HCl solution (7.2 mL, 43.2 mmol) was added and the mixture was stirred
at reflux temperature for 18 h. The solvent was evaporated at reduced
pressure and after that a 10% solution of K2CO3 was added to the residue until reaching pH = 10. The precipitate
obtained was filtered off and washed with H2O, affording
2.01 g of the product which was used for the following step without
further purification (77% yield); 1H NMR (400 MHz, DMSO-d6): δ 7.30 (br s, 2H), 7.78–7.91
(m, 1H), 7.95 (br s, 1H), 8.36–8.49 (m, 1H), 13.03 ppm (br
s, 1H); 13C NMR (101 MHz, DMSO-d6): δ 118.1, 118.6, 124.8, 140.7, 141.9, 150.5, 151.1 ppm; MS-ESI m/z: [M + H]+ calcd for C8H6ClN3, 180.0; found, 179.8; m/z: [M – H]− calcd
for C8H6ClN3, 178.0; found, 177.8;
HPLC (method 1): tR = 2.346 min.
Under an argon atmosphere, compound 69 (1.72 g, 9.61 mmol) was dissolved in dry DMF (20 mL) and
after cooling the reaction mixture to 0 °C, NaH (231 mg, 9.61
mmol) was added and the mixture was stirred at 0 °C for 15 min.
After that methyl iodide (1.61 mL, 25.6 mmol) was added dropwise and
the reaction mixture was let to heat to rt and stirred for 90 min.
The mixture was poured in H2O and the aqueous phase was
extracted three times with DCM. The combined organic layers were dried
over anhydrous Na2SO4 and the solvent was evaporated
at reduced pressure. Finally, the residue was treated with a mixture
of n-hexane/EtOAc 40:1 and the solid obtained was
filtered off and dried in vacuo, affording 793 mg of the product which
was used for the following step without further purification (43%
yield); 1H NMR (400 MHz, DMSO-d6): δ 3.82–3.93 (m, 3H), 7.09 (br s, 1H), 7.40 (br s,
1H), 7.71–7.78 (m, 1H), 7.80 (br s, 1H), 8.40–8.53 ppm
(m, 1H); 13C NMR (101 MHz, DMSO-d6): δ 34.8, 121.0, 121.6, 125.8, 128.7, 140.9, 142.3,
150.1, 150.8 ppm; MS-ESI m/z: [M
+ H]+ calcd for C9H8ClN3, 194.0; found, 193.8; HPLC (method 1): tR = 1.161 min.
Compound 72(19) (1.0 g, 4.29 mmol) was dissolved
in 30% HBr in AcOH (6 mL). After cooling the reaction mixture to 0
°C, Br2 (220 μL, 4.29 mmol) was added dropwise
and the reaction mixture was heated for 6 h at 40 °C. After evaporating
the solvent at reduced pressure, H2O was added and the
pH was adjusted to 8 using NH4OH solution. The water layer
was then extracted three times by DCM and the combined organic layers
were dried over anhydrous Na2SO4 and concentrated
at reduced pressure. Finally, the residue was purified by flash column
chromatography (SiO2, n-hexane/EtOAc 7:3),
affording 1.0 g of the desired compound (75% yield). Analytical data
were in agreement with the reported ones.[51]
1-(2-Chloropyridin-4-yl)propan-1-one (22)
(3.0 g, 17.68 mmol) was dissolved in a 30% solution of HBr in AcOH
(20 mL), and after cooling the mixture to 0 °C, bromine (900
μL, 17.68 mmol) was added and the reaction mixture was stirred
at 45 °C for 2 h and then heated to 75 °C and stirred for
additional 2 h. After evaporating the solvent at reduced pressure,
H2O was added and the pH was adjusted to 9 using NH4OH solution. The aqueous phase was extracted three times with
DCM and the combined organic layers were washed with H2O, dried over anhydrous Na2SO4, and concentrated
at reduced pressure. Finally, the residue was purified by flash column
chromatography (n-hexane/EtOAc 90:10 to 80:20), affording
2.3 g of the desired product (52% yield); 1H NMR (300 MHz,
CDCl3): δ 1.85 (d, J = 6.6 Hz, 3H),
5.07 (q, J = 6.6 Hz, 1H), 7.64 (dd, J = 5.1, 1.5 Hz, 1H), 7.76 (dd, J = 1.4, 0.7 Hz,
1H), 8.53 ppm (dd, J = 5.1, 0.7 Hz, 1H); 13C NMR (75 MHz, CDCl3): δ 19.5, 41.2, 120.4, 123.0,
143.2, 150.9, 152.9, 191.2 ppm; MS-ESI m/z: [M + H]+ calcd for C8H7BrClNO, 247.9; found, 248.0; m/z: [M + MeOH]+ calcd for C8H7BrClNO,
279.9; found, 280.0; HPLC (method 2): tR = 5.761 min.
Compound 76 (1.0 g, 4.0 mmol)
and N-methylthiourea (362.7 mg, 4.0 mmol) were dissolved
in EtOH (20 mL) and the reaction mixture was stirred at reflux temperature
for 1 h. The solvent was evaporated at reduced pressure and after
that the residue was suspended in H2O and the pH was adjusted
to 8 using NH4OH solution. The resulting suspension was
extracted three times with DCM and the combined organic layers were
washed with H2O and NaCl saturated solution, dried over
anhydrous Na2SO4, and concentrated at reduced
pressure. Finally, the residue was purified by flash column chromatography
(SiO2, DCM/EtOH 95:05) giving 510 mg of the desired product
(53% yield); 1H NMR (250 MHz, DMSO-d6): δ 2.43 (s, 3H), 2.83 (d, J = 4.9
Hz, 3H), 7.47 (q, J = 4.9 Hz, 1H), 7.61 (dd, J = 5.1, 1.5 Hz, 1H), 7.64–7.66 (m, 1H), 8.40 ppm
(dd, J = 5.1, 0.7 Hz, 1H); 13C NMR (101
MHz, DMSO-d6): δ 12.2, 30.5, 120.4,
121.2, 121.8, 141.0, 145.7, 149.8, 150.6, 165.5 ppm; MS-ESI m/z: [M + H]+ calcd, for C10H10ClN3S, 240.0;
found, 239.9; HPLC (method 2): tR = 9.091
min.
Authors: Véronique Plantevin Krenitsky; Lisa Nadolny; Mercedes Delgado; Leticia Ayala; Steven S Clareen; Robert Hilgraf; Ronald Albers; Sayee Hegde; Neil D'Sidocky; John Sapienza; Jonathan Wright; Meg McCarrick; Sogole Bahmanyar; Philip Chamberlain; Silvia L Delker; Jeff Muir; David Giegel; Li Xu; Maria Celeridad; Jeff Lachowitzer; Brydon Bennett; Mehran Moghaddam; Oleg Khatsenko; Jason Katz; Rachel Fan; April Bai; Yang Tang; Michael A Shirley; Brent Benish; Tracey Bodine; Kate Blease; Heather Raymon; Brian E Cathers; Yoshitaka Satoh Journal: Bioorg Med Chem Lett Date: 2011-12-10 Impact factor: 2.823
Authors: Laszlo Revesz; Ernst Blum; Franco E Di Padova; Thomas Buhl; Roland Feifel; Hermann Gram; Peter Hiestand; Ute Manning; Gerard Rucklin Journal: Bioorg Med Chem Lett Date: 2004-07-05 Impact factor: 2.823
Authors: Christoph Ernst; Johannes Heidrich; Catharina Sessler; Julia Sindlinger; Dirk Schwarzer; Pierre Koch; Frank M Boeckler Journal: Front Chem Date: 2018-10-16 Impact factor: 5.221